KR20180131775A - Method and System for detecting fire using genetic algorithm - Google Patents

Abstract

The present invention relates to a method for detecting fire using a genetic algorithm and a system thereof. The method comprises the following steps of: obtaining an image photographed by a thermal imaging camera and a sensing value detected by at least one sensor embedded in the thermal imaging camera; and applying a genetic algorithm to the obtained image and the at least one obtained sensing value to calculate a fire index so as to determine whether fire occurs. According to the method for detecting fire using a genetic algorithm and the system thereof, accuracy of determining whether fire occurs can be increased by detecting fire by not only a thermal image but also a sensor.

Description

FIELD OF THE INVENTION [0001] The present invention relates to a fire detection method and system using a genetic algorithm,

The present invention relates to a fire detection method and system using a genetic algorithm, and more particularly, to a fire detection method and system using a genetic algorithm capable of detecting fire occurrence using a genetic algorithm.

In the case of a fire detector installed on the ceiling of a structure to detect a fire, even if it is not an actual fire situation, it is often mistaken for a fire situation and malfunctions are often observed. These malfunctions are caused by moisture, sparks, smoke, etc., and the malfunctioning of firefighters and the evacuation of citizens are frequent.

According to the latest analysis of the number of fire alarms and fire alarms for 3 years in Gyeonggi Province, the number of fire alarms caused by the malfunction of fire alarms last year was 3,773, About 37 percent of the 11,414 cases were covered.

Japanese Patent Registration No. 10-1411624 discloses a method for detecting a fire through a visible light surveillance camera and a thermal imaging camera for fire detection. Specifically, a configuration is described in which a third image obtained by combining a first image photographed by a visible light monitoring camera and a second image photographed by a thermal image camera is generated, and a degree of fire is determined based on the hue of a specific entity from the third image .

However, since this method detects a fire by analyzing an image captured by a camera, there is a risk that the camera is visually obscured, malfunction may occur even if the temperature rises due to a cause other than a fire, or smoke is generated. There is an uneasy factor that interferes with

KR 101411624 B1

The present invention has been made in order to overcome the above problems, and it is an object of the present invention to provide a method and apparatus for detecting fire occurrence from a sensed value through an image captured through an infrared camera, The present invention relates to a fire detection method and system using a genetic algorithm.

According to another aspect of the present invention, there is provided a fire detection method using a genetic algorithm, the method comprising: obtaining a sensed value sensed by at least one sensor built in a thermal imaging camera and an image captured by a thermal imaging camera; And applying a genetic algorithm to the acquired image and at least one sensed value to determine a fire by calculating a fire index.

In the step of determining whether or not the fire is present, it is possible to determine whether or not the fire is caused by comparing preset reference data with an optimal solution to which the genetic algorithm is applied.

Preferably, the variable input to the genetic algorithm in the step of determining whether or not the fire is a fire may be a temperature value calculated based on the obtained image, and a sensing value detected by the at least one sensor including a smoke sensor or a gas sensor have.

In addition, the step of determining whether or not the fire has occurred may include calculating a temperature value calculated based on the obtained image using the genetic algorithm and a weight on the obtained sensing value.

The fire index may be calculated by comparing the calculated temperature value and a weight calculated based on the obtained sensing value with predetermined reference data.

In addition, the step of determining whether or not the fire has occurred may include transmitting the calculated fire index to the server.

The genetic algorithm can evolve the chromosome to be an optimal solution through generation of an initial population, fitness evaluation, selection, mating, and mutation operations.

The step of determining whether or not the fire has occurred may further include a step of informing a fire alarm when it is determined that a fire has occurred based on the calculated fire index or when information about a fire occurrence is received from the server .

A plurality of thermal cameras are provided, and fire detection of a dead zone, in which thermal imaging can not be performed by the first thermal imaging camera, is performed through a second thermal camera positioned at a predetermined interval on the diagonal line of the first thermal imaging camera .

In order to achieve the above object, a fire detection system using a genetic algorithm according to the present invention includes a thermal imaging camera including a control unit for determining a fire by calculating a fire index by applying a genetic algorithm; And a server for receiving the calculated fire index from the thermal imaging camera.

The thermal imaging camera comprises: a thermal imaging unit for imaging thermal images through an infrared camera and outputting the thermal imaging images to the control unit; A sensor unit having at least one of a smoke sensor and a gas sensor incorporated therein; And a transmission / reception unit for transmitting the calculated fire index to a server.

The controller may calculate a weight value based on the temperature value calculated based on the image obtained from the thermal image photographed by the thermal imaging unit and the sensing value detected by the sensor unit using the genetic algorithm.

If the transmitted fire index is determined to be an actual fire, the server transmits information about the occurrence of a fire to an infrared camera capable of communicating with the server in a certain range around the infrared camera transmitting the fire index .

The thermal imaging camera may include a fire alarm unit for generating a fire alarm when it is determined that a fire has occurred based on the calculated fire index or when information about a fire occurrence is received from the server.

According to the fire detection method and system using the genetic algorithm according to the present invention, since the fire is detected using not only the thermal image but also the sensor, the accuracy of fire occurrence determination is improved.

Also, since the value sensed through the sensor and the image photographed through the thermal camera learns the fire detection level which can be judged by fire through the genetic algorithm, the accuracy of the fire detection is improved.

1 is a block diagram briefly showing a fire detection system using a genetic algorithm according to the present invention,
Fig. 2 is a view for explaining an arrangement structure of the thermal imaging camera of Fig. 1,
3 is a flowchart showing a fire detection method using a genetic algorithm according to the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A fire detection method and system using a genetic algorithm according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram briefly showing a fire detection system using a genetic algorithm according to the present invention.

Referring to FIG. 1, a fire detection system 10 using a genetic algorithm according to the present invention includes a thermal imaging camera 100 and a server 200.

A plurality of thermal imaging cameras 100 may exist in a predetermined space. By positioning the second thermal imaging camera 100 on the diagonal portion of the first thermal imaging camera 100 located at the lower end of the drawing as shown in FIG. 2 , Fire detection can be performed by taking a thermal image through the second thermal imaging camera 100 in a dead zone where thermal imaging can not be performed by the first thermal imaging camera (100).

The thermal imaging camera 100 calculates a fire index by applying a genetic algorithm to determine whether or not the fire has occurred, and transmits the calculated fire index to the server.

The thermal imaging camera 100 includes a thermal imaging unit 110, a sensor unit 120, a control unit 130, a fire alarm unit 140, and a transceiver unit 150.

The thermal imaging unit 110 may be an infrared camera.

The thermal imaging unit 110 photographs the thermal imaging image through the infrared camera and outputs the thermal imaging image to the control unit 130.

The sensor unit 120 is mounted on the thermal imaging camera 100 and a temperature sensor 121, a gas sensor 122 and a smoke sensor 123 are applied.

The temperature sensor 121 may be omitted to detect the ambient temperature.

The control unit 130 uses the temperature value calculated based on the image obtained from the thermal image captured by the thermal imaging unit 110 and the sensed value detected by the sensor unit 120 using the genetic algorithm, .

The detailed processing procedure of the control unit 130 will be described later.

The fire alarm unit 140 is controlled by the control unit 130 to output a fire alarm, for example, an alarm sound or an alarm light.

The transmission / reception unit 150 is controlled by the control unit 130 and transmits the fire index calculated by the control unit 130 to the server 200.

In addition, the transceiver unit 150 receives the fire occurrence information from the server and provides it to the control unit 130.

The server 200 is connected to the thermal imaging camera 100 through wired or wireless data communication to transmit and receive information.

The server 200 receives and records information on the fire level from the plurality of thermal imagers 100 in a certain area.

In addition, when the server 200 receives a fire index determined as a fire, the server 200 is located in a certain vicinity of the thermal imaging camera 100 that transmitted the fire index, 200 to the infrared camera capable of communicating with the fire alarm system.

 Hereinafter, a fire detection process using a genetic algorithm will be described with reference to FIG.

First, thermal imaging and sensing values are obtained (step 310).

That is, the control unit 130 acquires the sensed value through the thermal image photographed by the thermal image sensing unit 110 and the sensor of the sensor unit 130.

Next, the fire index is calculated using the genetic algorithm (step 320).

The fire index calculation process using the genetic algorithm is performed in the controller 130.

That is, the temperature value calculated based on the thermal image taken by the thermal imaging unit 110 using the genetic algorithm in the control unit 130 and the sensed value detected through the sensor built in the sensor unit 130 Calculate the fire index.

The control unit 130 may be constructed to determine whether a fire has occurred by comparing preset reference data with an optimal solution to which a genetic algorithm is applied.

As an example, the control unit 130 may calculate a weight value based on the temperature value calculated based on the image acquired from the thermal imaging unit 110 and the sensing value acquired from the sensor unit 103 using the genetic algorithm , The temperature value and the sensed value can be compared with predetermined reference data to calculate the fire index.

Here, the predetermined reference data refers to information such as toxic gas concentration, smoke, temperature, and thermal camera image at the time of fire occurrence, and the fire index can be calculated by comparing the predetermined reference data with the weight value calculated by the genetic algorithm .

The fire index calculated through the genetic algorithm is an indicator for determining whether or not a fire has occurred. When the fire index calculated as a certain level or more is calculated, the controller 130 is constructed to determine that a fire has occurred.

Genetic algorithms are a process of natural evolution that evolves in a better way through changes in genes and an optimization algorithm that mimics the process of survival of the fittest, in which individuals who adapt well to the natural environment leave more offspring. The genetic algorithm is an algorithm that finds the optimal solution by expressing the solution to be searched by chromosome, generating a population, evaluating the fitness, and repeating mating, mutation and evaluation for the chromosomes in the selected population.

The chromosome of the genetic algorithm according to an embodiment of the present invention may correspond to the extracted thermographic image and the extracted sensing value. In addition, in this case, the chromosome may be extracted from the extracted thermal image and the sensing of each sensor built in the sensor unit 130. In this case, Value. ≪ / RTI >

The control unit 130 can be constructed to learn the fire level using the extracted thermal image and a genetic algorithm using the extracted sensing value as a chromosome, and to determine the fire level in the current situation.

That is, the control unit 130 calculates the weights of the temperature and the sensing value by calculating the temperature based on the temperature calculated based on the thermal image imaged by the thermal imaging unit 110 and the sensing value extracted by the sensor unit 130 The fire level can be determined with comprehensive consideration.

Because fire level is learned and the fire level for each situation is determined by taking into consideration the calculated weight and the calculated weight in total, it is necessary to use only the infrared camera, It is possible to improve the accuracy in judging the level of fire occurrence.

The control unit 130 can generate a population of chromosomes and evaluate the fitness for each chromosome in the population. A chromosome with a high evaluation value can be selected, crossing and mutation manipulation can be applied to each chromosome, and the fitness can be evaluated to generate a population of the next generation. When the intersection, the mutation operation, and the fitness evaluation are repeated and the optimal solution is derived, the controller 130 can determine the fire level by calculating the fire index for the current situation.

Thereafter, the calculated fire index is transmitted to the server 200 (step 330), and it is determined whether the calculated fire index corresponds to a fire occurrence (step 340).

The determination as to whether or not a fire occurs based on the fire index is made by comparing the fire index calculated by the controller 130 with an index corresponding to the occurrence of fire as described above.

If it is determined in step 340 that the fire does not correspond to the fire occurrence, the server 200 determines whether the fire occurrence information corresponding to the fire detection from the other thermal imaging camera is received (step 360).

If it is determined in step 340 that the fire is generated, or if it is determined in step 360 that the fire occurrence information is received from the server 200, the fire alarm is generated through the fog alarm unit 140 (step 350).

According to the fire detection method and system using the genetic algorithm described above, since the fire is detected by using the sensor as well as the thermal image, the accuracy of fire occurrence determination is improved.

Also, since the value sensed through the sensor and the image photographed through the thermal camera learns the fire detection level which can be judged by fire through the genetic algorithm, the accuracy of the fire detection is improved.

100: Thermal camera
110: thermal imaging unit
120:
130:
140: Fire alarm
150: Transmitting /
200: Server

Claims (16)

Obtaining a sensed value detected by at least one sensor incorporated in the thermal imaging camera and an image photographed by the thermal imaging camera;
And generating a fire index by applying a genetic algorithm to the acquired image and at least one sensed value to determine whether or not the fire is present. The method according to claim 1,
The step of determining whether or not the fire
And comparing the preset reference data with an optimal solution to which the genetic algorithm is applied to judge whether or not a fire is present. The method according to claim 1,
In the step of judging whether or not the fire has occurred
Wherein a variable input to the genetic algorithm is a sensed value detected by the at least one sensor including a smoke sensor or a gas sensor and a temperature value calculated based on the obtained image. . The method according to claim 1,
The step of determining whether or not the fire
Calculating a temperature value calculated based on the obtained image using the genetic algorithm and a weight on the obtained sensing value. 5. The method of claim 4,
The fire index
And comparing the computed temperature value and a weight calculated based on the obtained sensing value with preset reference data. The method according to claim 1,
The step of determining whether or not the fire
And transmitting the calculated fire index to a server. The method according to claim 1,
The genetic algorithm
Wherein the evolution of the chromosome to an optimal solution is applied through generation of an initial population, evaluation of fitness, selection, mating, and mutation operations. The method according to claim 6,
The step of determining whether or not the fire
If it is determined that a fire has occurred based on the calculated fire index or
Further comprising the step of generating a fire alarm when information on the occurrence of a fire is received from the server. The method according to claim 1,
The thermal imaging camera has a plurality of
Wherein a fire alarm is detected in a dead zone where thermal imaging can not be performed by a first thermal camera through a second thermal camera positioned at a predetermined interval on a diagonal line of the first thermal camera. Fire detection method. An infrared camera including a control unit for determining whether a fire is caused by calculating a fire index by applying a genetic algorithm; And
And a server for receiving the calculated fire index from the thermal imaging camera. 11. The apparatus of claim 10, wherein the thermal imaging camera
A thermal image pickup unit for taking a thermal image through an infrared camera and outputting the thermal image to the control unit;
A sensor unit having at least one of a smoke sensor and a gas sensor incorporated therein; And
And a transmission / reception unit for transmitting the calculated fire index to a server. 12. The apparatus of claim 11, wherein the control unit
And a weight value is calculated based on a temperature value calculated based on the image obtained from the thermal image photographed by the thermal imaging unit and a sensing value detected by the sensor unit using the genetic algorithm. Fire detection system. 13. The apparatus according to claim 12,
Wherein the fire index is calculated by comparing the calculated temperature value and a weight calculated based on the detected sensing value with predetermined reference data. 11. The system of claim 10, wherein the server
And transmits information on fire occurrence to a thermal imaging camera capable of communicating with the server located in a certain range around the thermal imaging camera transmitting the fire index when it is determined that the fire index transmitted is a real fire Fire detection system using genetic algorithm. 11. The apparatus of claim 10, wherein the thermal imaging camera
And a fire alarm unit for generating a fire alarm when it is determined that a fire has occurred based on the calculated fire index or when information about a fire occurrence is received from the server. . 11. The system of claim 10, wherein a plurality of thermal imaging cameras are applied,
Wherein a fire alarm is detected in a dead zone where thermal imaging can not be performed by a first thermal camera through a second thermal camera positioned at a predetermined interval on a diagonal line of the first thermal camera. Fire detection system.

Images